Background & Timeline

• Starting in 1994, “Stop, Question & Frisk” is a strategy resulting from “Broken Window Policing”.
• On January 31st, 2008, a federal class action lawsuit claimed Stop & Frisk “encourages racial profiling and unconstitutional detainment”.
• On May 16th, 2012, a judge granted class certification for the lawsuit.
• The trial begins on March 18th, 2013 and ends on August 12th, 2013 with a judge ruling Stop & Frisk to be unconstitutional.
• Stop & Frisk still occurs in New York City today.

Literature Review

Other impacts of Stop & Frisk:

Decreased educational outcomes (Legewie and Fagan 2019; Bacher-Hicks and Campa 2021a)

Decreased views of police legitimacy (Tyler, Fagan, and Geller 2014)

Decreased mental health (Geller et al. 2014)

Housing Prices:

(Friedman 2015) shows that properties exposed to more intense Stop & Frisk behavior sell for lower prices

Each additional stop decreases prices by $35 - $250

Using data from 2006 - 2012 (includes recession & avoids court case)

Does not control for crime

Data

1. New York City (Residential) Property Sales
   • NYC Dept. of Finance
   • Automated City Register Information System (ACRIS)
2. New York Police Department (NYPD)
   • NYPD Operation Impact Zones\(^{*}\)
   • NYPD Stop & Frisk and Crime Data
3. American Community Survey Data for New York City
4. Census Bureau 2000 and 2010 Decennial Census Surnames

Methodology

I want to know how property prices changes after Stop & Frisk was ruled unconstitutional.

Methodology

I want to know how property prices changes after Stop & Frisk was ruled unconstitutional.

Methodology

I want to know how property prices changes after Stop & Frisk was ruled unconstitutional.

Methodology

I want to know how property prices changes after Stop & Frisk was ruled unconstitutional.

Model 1

\[\begin{align}P_{int} &= X_{int}\beta + \delta_1 \text{Floyd}_{t} + \delta_2 \text{Impact}_{i} \\ &+ \delta_3 (\text{Floyd}_{t} \times \text{Impact}_{i}) + \epsilon_{int}\end{align}\]

where:

\(X_{int}\) are variables specific to the property.

Floyd\(_{t}\) is an indicator equal to 1 for periods between trial dates.

Impact\(_i\) is an indicator equal to 1 for properties inside Impact Zones.

Model 2

\[\begin{align}P_{int} & = X_{int}\beta + \delta_1 \text{Floyd}_{t} + \delta_2 \text{Impact}_{i} \\ & + \delta_3 (\text{Floyd}_{t}\times \text{Impact}_{i}) + \lambda_1 \text{NonWhite}_{it} \\ & + \lambda_2 (\text{Impact}_{i}\times \text{NonWhite}_{it}) + \lambda_3 (\text{Floyd}_{t}\times \text{NonWhite}_{it}) \\ & + \lambda_4 (\text{Floyd}_{t}\times \text{Impact}_{i}\times \text{NonWhite}_{it}) + \epsilon_{int}\end{align}\]

where:

NonWhite is an indicator if the last name of the buyer/seller has a \(<50\)% chance of being white. \(\lambda_4 > 0\) would mean that treated non-white individuals saw a greater price increase than their white counterparts.

Model 2

\[\begin{align} P_{int} &= X_{int}\beta + \delta_1 \text{Floyd}_{t} + \delta_2 \text{Impact}_{i} \\ & + \delta_3 (\text{Floyd}_{t} \times \text{Impact}_{i}) + \gamma_1 (\text{Impact}_{i} \times \text{Exposure}_{it}) \\ & + \gamma_2 (\text{Floyd}_{t} \times \text{Impact}_{i} \times \text{Exposure}_{it}) + \epsilon_{int}\end{align}\]

where:

Exposure is an indicator if the seller owned the property while it was in an active Impact Zone. \(\gamma_2 > 0\) would mean that treated individuals who lived through an Impact Zone saw greater price increases than their counterparts who did not directly experience an Impact Zone.

Conclusion

1. Contrary to Friedman (2015), the intensive margin of Stop & Frisk itself does not appear to influence housing prices
2. Values were previously depressed in areas subjected to Operation Impact before Stop & Frisk was ruled to be illegal
3. The bounce-back in housing prices was not realized by those people who were most heavily subjected to Stop & Frisk beforehand.
   • The literature has documented negative unintended consequences resulting from Stop & Frisk (e.g. education, mental health).
   • The effect Stop & Frisk had on housing prices was about equal to that of an economic recession.

Other Work in Progress